Abstarct: Perhaps the most fundamental concepts in the field of statistic is the porobability function which can be found in the behavior of random variables especially in statistics without knowing the statistical probability of a random sample‎, ‎ability to perform inference on random samples and generalizing the results to the community does not exist‎. ‎On of the methods for estimating density (probability) function is‎ "‎kernel density estimation"‎, ‎the generalization of naive estimator‎. ‎In kernel methods‎, ‎we must specify two parameters‎: ‎Kernel function and bandwidth‎. ‎The bandwidth selection plays a key role in kernel methods‎. ‎When bandwith is small‎, ‎the curve is nonsmooth‎. ‎With increasing the bandwith‎, ‎the smoothness increases‎. ‎Some of the methods for bandwidth selection is‎: ‎subjective methds‎, ‎normal scale rule‎, ‎least square cross validation etc‎. ‎In this disseration‎, ‎we study density estimation via kernel methods‎, ‎also‎, ‎compare some of the bandwidth selection methods and apply them to astrostatistics data‎.